三维激光扫描在地形监测中的应用

文中首先采用 Leica Scanstation2 3D激光扫描系统进行了野外现场数据的收集，基于收集的实时数据对点云数据进行了拼接、去噪、过滤等处理，获取了高准确率的点云数据。之后，将点云数据输入 Matlab中，再采用 BP神经网络方法进行曲面重构，在不同的时间内通过对同一地区所得的曲面拟合进行分析，并控制其精度、观察其形变范围、计算其变形程度，以便对形变区的形变趋势进行实时监控与预测。

Application of Three-dimensional Laser Scanning in Terrain Monitoring

In this paper, the Leica Scanstation2 3D laser scanning system is used to collect field data. Based on the collected real-time data, the point cloud data is spliced, denoised, filtered and other processes to obtain high-accuracy point cloud data. After that, the point cloud data is input into Matlab, and then the BP neural networks method is used to reconstruct the surface. The surface fitting obtained in the same area is analyzed at different times, and its accuracy is controlled, its deformation range is observed, and its deformation degree is calculated, so as to monitor and predict the deformation trend of the deformation area in real time.